局部序列比对算法及其并行加速研究进展

Advances in local sequence alignment algorithm and its parallel acceleration research

随着新一代测序技术的发展,传统的序列比对工具已无法满足测序产生的海量生物学数据分析处理的需求,研究如何利用最新的计算技术加速序列比对过程具有十分重要的意义。本文回顾了常用的局部序列比对算法,介绍了基于并行计算原理的序列比对算法的加速优化策略和主要进展,详细说明了如何利用最新的图形处理器(GPU)计算技术实现高性能的BLAST(basic local alignment search tool)比对算法。最后,结合实际需求,提出和讨论了综合利用云计算和GPU计算实现高性能、高能效的序列比对平台的研究思路。

With the development of next-generation sequencing technology, the traditional sequence alignment tools can no longer meet the need to analyze these vast amounts of biological data generated by the new sequencing technology. So it is of great significance to study how to accelerate the alignment process with the latest computing technology. In this article a brief review of the common local alignment algorithms is given first, and then optimization strategies and main research advances in sequence alignment algorithms based on parallel computing are introduced. After that, a detailed description of realizing high-performance basic local alignment search tool （BLAST） algorithm with the latest graphics processing unit （GPU） computing technology is given. According to the actual need, a high-performance, energy-efficient sequence align- ment platform using cloud computing and GPU computing is suggested and discussed.